Load XGBoost model from path in HDFS-compatible file system
Load XGBoost model from path in HDFS-compatible file system
The path of the file representing the model
The loaded model
Train XGBoost model with the DataFrame-represented data
Train XGBoost model with the DataFrame-represented data
the training set represented as DataFrame
Map containing the parameters to configure XGBoost
the number of iterations
the number of xgboost workers, 0 by default which means that the number of workers equals to the partition number of trainingData RDD
An instance of ObjectiveTrait specifying a custom objective, null by default
An instance of EvalTrait specifying a custom evaluation metric, null by default
indicate whether to use external memory cache, by setting this flag as true, the user may save the RAM cost for running XGBoost within Spark
The value which represents a missing value in the dataset
the name of input column, "features" as default value
the name of output column, "label" as default value
XGBoostModel when successful training
ml.dmlc.xgboost4j.java.XGBoostError
when the model training is failed
Train XGBoost model with the RDD-represented data
Train XGBoost model with the RDD-represented data
the training set represented as RDD
Map containing the configuration entries
the number of iterations
the number of xgboost workers, 0 by default which means that the number of workers equals to the partition number of trainingData RDD
An instance of ObjectiveTrait specifying a custom objective, null by default
An instance of EvalTrait specifying a custom evaluation metric, null by default
indicate whether to use external memory cache, by setting this flag as true, the user may save the RAM cost for running XGBoost within Spark
The value which represents a missing value in the dataset
XGBoostModel when successful training
ml.dmlc.xgboost4j.java.XGBoostError
when the model training has failed
Train XGBoost model with the RDD-represented data
Train XGBoost model with the RDD-represented data
the training set represented as RDD
Map containing the configuration entries
the number of iterations
the number of xgboost workers, 0 by default which means that the number of workers equals to the partition number of trainingData RDD
An instance of ObjectiveTrait specifying a custom objective, null by default
An instance of EvalTrait specifying a custom evaluation metric, null by default
indicate whether to use external memory cache, by setting this flag as true, the user may save the RAM cost for running XGBoost within Spark
the value represented the missing value in the dataset
XGBoostModel when successful training
Use XGBoost.trainWithRDD instead.
ml.dmlc.xgboost4j.java.XGBoostError
when the model training is failed